exercise data (only headers - redacted for legal reasons):



App_Users_in_Graph_demographics.txt: is the demographic information of the running tracker users that have at least one running-buddy (network-embedded). Demographics include user unique identifier, year of birth, gender, weight, height, join date, device type and the address at the time of registration that includes city, state, country and ZIP code (or post-code). We extract the exact geographic location of users in Latitude and Longitude either through the running GPS information in data file GPS_summary or in the absence of GPS tracking through the address information using the Google Maps API (https://developers.google.com/maps/).



SPA_Social_graph.txt: is the running activity records of the network-embedded individuals. Each exercise activity is a row in the dataset. Includes information about the user id of the runner, the type of activity (if it is run, walk etc.), the duration of the activity, the distance, the calories burned, the local start time of the event as well as the timezone of the event. Datenumber of the event refers to number of days that the event took place calculated from January 0, 0000 (Matlab�s specific way to transform dates into numbers). 



GPS_summary.txt: gives the exact starting geographic location in Latitude and Longitude of an exercise activity. Information includes the activity unique identifier, the provider of the GPS tracking (creator), the starting LAT and LON as well as the user unique identifier. We use this data to infer the geographic location of the users. 



USERREL.txt: is the social network of the individuals (running-buddy network). Includes information about the two individuals that form a link as well as a timestamp for the date that the link was created.





 






